Mirror MaStR database by continously querying latest data

[gplssm]

Because downloading the entire MaStR database from scratch ever time an update is needed takes too long, a continuous updater is required that more or less mirrors the MaStR database.

This will be implemented using a (dockered) PostgreSQL database access through SQLAlchemy. It is checked for new units or units with updated data which is then pull into the database. In a addition, extended data about the unit is requested for download. The list of extended data requests gets iterated in a subsequent process and data is retrieved.

Tasks

• Download function for basic unit data
• Basic implementation of extended unit download (prototyping with wind)
• Dump and restore functionality
• ORMs of extended unit data
  • wind
  • solar
  • biomass
  • nuclear
  • combustion
  • hydro
  • gsgk
  • storage
  • consumer _postponed to Allow for downloading additional data for consumer and gas sector units #163
  • gas_consumer _postponed to Allow for downloading additional data for consumer and gas sector units #163
  • gas_producer _postponed to Allow for downloading additional data for consumer and gas sector units #163
  • gas_storage _postponed to Allow for downloading additional data for consumer and gas sector units #163
• ORMs of eeg unit data
  • wind
  • solar
  • biomass
  • hydro
  • gsgk
  • storage
• ORMs of kwk unit data
  • biomass
  • combustion
  • gsgk
• ORMs of permit unit data
  • wind
  • solar
  • biomass
  • nuclear
  • combustion
  • hydro
  • gsgk
  • storage
• Configure backfill_basic() such that only data with newer timestamp gets inserted, see here
• Remove data dependent column creation when flattening in flatten_rule_expand
• Re-include adopted time blacklisted that can control number of prcesses used to download additional unit data maybe later Consider to re-include blacklisted time  #148
• Properly delete successful additional data download requests not soo urgent anymore, postponed Make sure SQLAlchemy properly deletes successful when additional data is downloaded #154
• Allow multiprocessing with basic_unit_data() postponed Parallel download in basic_unit_data() #161
• Download only EEG, KWK and permit data for units with "StatistikFlag=='B'". This information is included in extended unit data, see todo note here
• Add function to create new additional data requests for units with EegMastrNummer, KwkMastrNummer, and so on and recreate requests after inital backfill was completed
• Delete additional data requests for units where it failed X times
• Make private methods of download.py that are used during mirroring public
• Sort code to desired locations and maybe rename files
• CSV export to *_raw.csv compatible format
• Handle NewConnectionError type postponed NewConnectionError: Temporary failure in name resolution #155
• Write docs and link class
• Add to CHANGELOG